Analytics Everywhere: Generating Insights From the Internet of Things

Abstract

The Internet of Things is expected to generate an unprecedented number of unbounded data streams that will produce a paradigm shift when it comes to data analytics. We are moving away from performing analytics in a public or private cloud to performing analytics locally at the fog and edge resources. In this paper, we propose a network of tasks utilizing edge, fog, and cloud computing that are designed to support an Analytics Everywhere framework. The aim is to integrate a variety of computational resources and analytical capabilities according to a data life-cycle. We demonstrate the proposed framework using an application in smart transit. INDEX TERMS Descriptive analytics, diagnostic analytics, predictive analytics, edge computing, fog computing, cloud computing, Internet of Things.

DaLiF: a data lifecycle framework for data-driven governments

“…The other existing data lifecycles include: Data Lifecycle for HPC Scientific data perspective [114], Data lifecycle for cloud automation tools [40], Data Lifecycle for Telco networks data management [115], Energy big data lifecycle [116], Data lifecycle for the Tobacco industry [117], Data lifecycle for cloud computing [118];Data lifecycle for cloud data [119], Data lifecycle for IoTs [120], Personal data lifecycle [121], Data lifecycle about the coal mine industry [122], Data lifecycle for smart healthcare [123], Data Lifecycle Model for NSF [94], Data lifecycle cycle for smart cities [13], Storage data lifecycle [124], Research data lifecycle [125], lifecycle for CENS Data [126], data lifecycles for industry [127,128], a lifecycle for big scholarly data [129], a lifecycle for social and economic data [46], Data lifecycle for manufacturing [130], Research data lifecycle [131], a lifecycle for big healthcare data [23,132], data lifecycle [133], a lifecycle for environmental research data [134], a lifecycle for big data value creation [26], a lifecycle for big data analytics for psychologists [135], the information pyramid of Reynolds and Busby lifecycle [17], Yuri Demchenko data lifecycle [10], Data lifecycle [58], SCC-data lifecycle [44,54], Data value cycle [136], Lifecycle in databases [137], Knowledge process-lifecycle [138], CMM for Scienti...…”

A Holistic Overview of Anticipatory Learning for the Internet of Moving Things: Research Challenges and Opportunities

“…Many context models will require simple machine learning algorithms such as the linear Spanish inquisition protocol (L-SIP) which has been applied to reduce data transmission; filtered state classification (ClassAct) as a human posture/activity classifier based on decision tree; and time-discounted histogram encoding (Bare Necessities) which is used for summarizing the relative time spent in given contexts [94]; • Mobility and geographic distribution: These are indispensable requirements for context intelligence; however, an anticipatory learning system also requires a rich scenario of communication and interaction between all available computational resources. To achieve this, a priori data pipelines must be designed that will support an analytics everywhere framework [95][96][97]; • Heterogeneity and interoperability: Obviously, terminal devices in the IoMT system can collect data with different timestamps, formats, and locations. Additionally, the edge network computing devices which deploy the IoT gateways could seamlessly support the interoperability between terminal devices.…”

An Edge-Fog-Cloud Architecture of Streaming Analytics for Internet of Things Applications

“…Our proposed architecture is a step forward in finding a unique solution that combines the advantages of different computational resources into an integrated edge-fog-cloud fabric that is capable of capturing, managing, processing, analyzing and visualizing IoT data streams. This fabric of computational resources is designed to work towards an asynchronous approach for supporting an Analytics Everywhere framework [19] making the development, deployment and maintenance more pragmatic and scalable. By breaking down the processing and analytical capabilities into a network of streaming tasks and distributing them into an edge-fog-cloud computing environment, our proposed architecture can support streaming descriptive, diagnostic and predictive analytics.…”

“…In general, an IoT application will require a combination of different compute nodes running at the edge, fog and/or cloud. The main criteria to take into account when selecting a compute node have been first introduced by Cao and Wachowicz [19]. They are described as follows:Vicinity: The geographical proximity of compute nodes to an IoT device is an important criterion to take into consideration for an IoT application.…”